Visualization 1

Column

Version 1

This was the first attempt I made in developing this plot. I quickly realized that Group needed to be recoded as a factor. I also noticed that the labels for the X Axis are redundanct and would be better as numbers only. So in my next iteration, I will remove the “RSA_” from my axis labels and I will recode Group as a factor.

Version 2

After making these changes I saw improvements, but there was also another obvious problem that popped up - no data showing. I sought out some help for this and discovered that I was mapping things in the aesthetics wrong. I needed to define “group” rather than color is the main plot aesthetic and define color in the geom.

Version 3

Wahoo! Data has appeared. At this point I feel like the visual is fairly successful, but can use some little tweaks to the design. In my next batch of edits, I’ll be defining the groups more clearly and adding some style changes.

Column

Final Version

For my final edits, I improved a lot of basic visual aspects of the plot - line width, color, text size, title centering etc. I also incorporated edits from my peer review which suggested showing some uncertainty. I created two versions of the final visual - 1 with the full y-axis and one zoomed in. This plot shows the entire y-axis. You’ll notice there is a lot of white space here because there are no data points in the lower range, which is not visually appealing. The benefit of this however, is that it more accurately shows the scale of the axis and doesn’t exaggerate differences between groups.

Final Version Zoomed In

In this version of the finalized plot, I’ve limited the y-axis to show only a select range. This allows you to see the group differences more clearly. A disadvantage to this, is that it may exxagerate differences, making things appear significantly different even though they are actually quite close. Because this data is visualizing HRV however, it makes sense to limit the range as there are physiological limitations to how the scale extends (only dead people would have an HRV at 0). Also differences in HRv can be small but quite meaninful at times.

Visualization 2

Version 1

This was my first attempt at this visual. A pretty decent start. I recognized the need for effective labelling and went ahead and did that right away so that the axes labels, factors, and group are all easily determinable. While I think this is a good first attempt, I wanted to improve some of the aesthetics and I worked on that in my next iteration.

Version 2

For this version, I worked on improving some visual aesthetics - changing the color scale, adjusting transparency, centering the title, and changing the panel background. Also at the recommendation of one of peer reviews I created a line at the mean to highlight that more clearly in the plot.

Final Version

In this final version I added error bars to visualize uncertainty. This seems effective however one of my error bars is not showing up and I can see that is because one of the standard erros is “NA”. Not sure what created this or how to fix it at this point.

Visualization 3

Column

Version 1

Version 2